---
license: apache-2.0
pipeline_tag: text-to-speech
library_name: gguf
base_model: Zyphra/ZONOS2
base_model_relation: quantized
tags:
  - text-to-speech
  - tts
  - gguf
  - ggml
  - zonos2.cpp
---

# ZONOS2 — GGUF

<p align="center">
  <img src="https://huggingface.co/Zyphra/ZONOS2/resolve/main/assets/ZONOS2BlogThumbnail.png" alt="ZONOS2 title card" width="750" />
</p>

<div align="center">
  <a href="https://discord.gg/gTW9JwST8q" target="_blank">
    <img src="https://img.shields.io/badge/Join%20Our%20Discord-7289DA?style=for-the-badge&logo=discord&logoColor=white" alt="Discord">
  </a>
</div>

---

GGUF weights for [**Zyphra/ZONOS2**](https://huggingface.co/Zyphra/ZONOS2), ready to run
with [**zonos2.cpp**](https://github.com/Zyphra/zonos2.cpp) — a standalone ggml/GGUF C++
port of ZONOS2. The **entire pipeline** (ECAPA speaker encoder → MoE backbone → DAC
vocoder) runs as native C++ linking only `libggml` + `gguf`: **no Python, no PyTorch, no
CUDA-only kernels at inference time.** CPU and CUDA use the same files.

ZONOS2 is Zyphra's latest text-to-speech model, trained on more than 6 million hours of
varied multilingual speech, delivering expressiveness and quality on par with—or even
surpassing—top TTS providers at low latency with MoE. ZONOS2 excels at high-fidelity and
naturalistic voice cloning.

**For model details and speech samples, check out our [blog](https://www.zyphra.com/our-work/zonos2).**
**A hosted version is available at [cloud.zyphra.com/audio-playground](https://cloud.zyphra.com/audio-playground).**

Language support is as follows.
| Tier   | Languages                                                                                                                                                                                      |
| ------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Tier 1 | English, Mandarin Chinese, Japanese                                                                                                                                                            |
| Tier 2 | Korean, Russian, Italian, Portuguese, French, Spanish, Vietnamese, German, Hebrew, Dutch                                                                                                       |
| Tier 3 | Swedish, Hindi, Tamil, Telugu, Thai, Norwegian, Bengali, Tagalog, Arabic, Danish, Indonesian, Polish, Ukrainian, Romanian, Finnish, Hungarian, Lithuanian, Estonian, Slovak, Croatian, Latvian |

---

## Files

| File | Size | Description |
| --- | --- | --- |
| `zonos2-f16.gguf` | 15.3 GB | F16 backbone — lossless from the bf16 checkpoint |
| `zonos2-q8_0.gguf` | 8.5 GB | Q8_0 MoE experts, **F16 spine** — recommended; effectively lossless |
| `zonos2-q6_k.gguf` | 6.8 GB | Q6_K MoE experts, F16 spine |
| `zonos2-q5_k.gguf` | 5.9 GB | Q5_K MoE experts, F16 spine |
| `zonos2-q4_k.gguf` | 4.9 GB | Q4_K MoE experts, F16 spine — smallest |
| `dac.gguf` | 254 MB | DAC-44 kHz decoder (codes → waveform) |
| `spk-encoder.gguf` | 24 MB | ECAPA-TDNN speaker encoder (wav → x-vector, for voice cloning) |

All quants keep the **spine (attention + router + embeddings + speaker projection) at
F16** and quantize only the MoE expert matrices — the layout that holds quality far below
the usual quant floor. The full-precision spine is the single biggest quality lever; pick
the expert precision (Q8 → Q4) that fits your VRAM.

## Benchmarks

| Build | Size | bpw | KLD&nbsp;↓ | Top-1&nbsp;↑ | WER&nbsp;↓ | SpkSim&nbsp;↑ | UTMOS&nbsp;↑ |
| --- | --- | --- | --- | --- | --- | --- | --- |
| **F16** (ref) | 15.3 GB | 16.0 | — | — | 2.79 | 66.75 | 4.40 |
| **Q8_0** | 8.5 GB | 8.50 | 0.002 | 96.5% | 2.87 | 66.30 | 4.40 |
| **Q6_K** | 6.8 GB | 6.56 | 0.007 | 92.9% | 3.07 | 66.12 | 4.40 |
| **Q5_K** | 5.8 GB | 5.50 | 0.025 | 86.3% | 2.98 | 66.30 | 4.40 |
| **Q4_K** | 4.9 GB | 4.50 | 0.072 | 76.9% | 3.00 | 64.54 | 4.36 |

- **KLD** (mean) and **Top-1** measure how closely each quant tracks the F16 backbone's
  per-frame logits, scored with the `zonos2-perplexity` tool in zonos2.cpp.
- **WER** (Qwen3-ASR word-error rate), **SpkSim** (clone speaker similarity), and
  **UTMOS** (predicted MOS) are end-to-end on the
  [`Zyphra/ZTT1-Eval`](https://github.com/Zyphra/ZTT1-Eval) Clean English set;
  lower WER and higher SpkSim/UTMOS are better.

Although the logit metrics (KLD, Top-1) degrade steadily as the experts shrink, the
**audio quality holds nearly flat down to Q4_K** — WER, speaker similarity, and UTMOS stay
within eval noise of F16. The F16 spine keeps the model on-distribution, so the smaller
quants spend their error budget on inaudible logit jitter rather than audible artifacts.
**Q8_0 is the recommended default** (effectively lossless); Q4_K is a strong choice when
VRAM is tight.

## Quick Start

Build [zonos2.cpp](https://github.com/Zyphra/zonos2.cpp) (CPU, or `-DGGML_CUDA=ON` for
NVIDIA), then turn text into a waveform with one command:

```bash
zonos2-cli zonos2-q8_0.gguf --tts "Hello, world." out.wav \
    --dac dac.gguf --gpu --seed 1
```

Add `--spk voice.mp3` (with `spk-encoder.gguf`) to clone a voice from a reference clip.

### HTTP server

`zonos2-server` mirrors the reference FastAPI — low-latency streaming PCM, an OpenAI
`/v1/audio/speech` endpoint, in-process voice cloning, and a browser UI:

```bash
zonos2-server zonos2-q8_0.gguf --dac dac.gguf --spk-encoder spk-encoder.gguf --gpu
```

See the [zonos2.cpp README](https://github.com/Zyphra/zonos2.cpp) for build instructions,
quantization (`quantize-cli`), batching, and the full CLI reference.

## Citation
If you find this model useful in an academic context please cite as:
```
@misc{zyphra2025zonos,
  title     = {Zonos V2 Technical Report},
  author    = {Gabriel Clark, Sofian Mejjoute, Mohamed Osman, George Close, Beren Millidge},
  year      = {2026},
}
```